GoodSoundClub - Romy the Cat's Audio Site

In the Forum: Horn-Loaded Speakers
In the Thread: Macondo’s Midbass Project – the grown up time.

Post Subject: Bumps & springs & exciting thingsPosted by jessie.dazzle on: 10/12/2010

Romy wrote:

"... from the perspective of “reading the back side of the midbass speed bump” I would argue that it has nothing to do with horn loading. Look at the notion widely. I would propose the following..."

"...An abstract ultimate playback system MUST have a Midbass Channel that is driven by an amplifier with negative output impedance. The output impedance of the amp shall be equal to the impedance of the driver and the impedance of the amp shall cancel-out all mechanical influences that the driver diaphragm experience..."

Well I would like to consider this view, but I do not fully understand the relationship between output impedance and diaphragm control; also I don't know what you mean by "negative impedance". Your mid-bass drivers have 15 Ohm coils; from what you write, I assume you are driving them from 15 or 16 Ohm taps. Correct?

To answer Paul... (Hm, I'd better sit down):

Our favorite intersection indeed! I'll be moving about 6K miles closer to you, so maybe in the future you can just stop by the house (part of my crusade to get you back into horns!).

Yes, in theory, what I say above should also apply to ULF, but in practice, well just calculate the dimensions for an 18Hz exponential horn.

Horns come with their own horn problems, mostly when asking too much of a single horn; otherwise the problem is mainly one of cost and space, and the cost of that space.

Regarding the lack of a linear load resulting from compressibility of the medium (the air): The job of a diaphragm is to excite, not "pump" the air (this cannot be overstated... It is probably the single biggest misconception in audio). It is easier to excite a spring than a tree stump. Like a tuning fork, it is the "springiness" of the air that allows it to be excited.

At some point, whether direct-radiator or horn-loaded, the excited air directly in front of the diaphragm must couple with and excite the still air in the room. A direct radiator flaps away with its hard surface acting directly on the low pressure air in the room; because the air is not confined, the transfer of energy is extremely inefficient; ever try paddling a boat with your fingertips? In either case, you are mostly just punching holes. The throat of a horn confines the air on all but one side; when the diaphragm moves, the result is a high-pressure/low area condition, which the flare converts to increasingly lower pressure acting over an increasingly larger area. At the same time, the pulses are conveyed (accelerated and decelerated) in a medium that is compliant (compressible), which thereby momentarily exaggerates the pulses (both positive and negative pressures); the "spring" momentarily stores then releases energy but still must keep time, so the energy ends up being released with more force. When the pulses finally reach the mouth and interact with the air in the room, they are characterized by low pressure differentials, acting over a very large area, via a compliant medium. Both of these factors contribute to better "grip" on the still air (the uptake is progressive; ever try to push a soap bubble with a quick slap?). Rather than punch a hole and simply create turbulence, the air in the room is now more precisely or accurately excited.

Regarding linearity of load: I assume you are referring to the direct-radiator firing straight into open space and consequently seeing a more linear air load... It is precisely the inability of direct-radiators to efficiently couple with their load that makes them not "see" the load; so yes, what they see is more linear; it is the linear absence of a load. Again, the load they do see is mainly the inertia of the cone. One might assume that the enclosure, provides some back pressure, and therefore a load, but think about it; the diaphragm moves a few millimeters against a volume of several liters, which most often turns out to be ported... This is not going to create a load anywhere near what it takes to rocket that cone back and forth twice as fast (see previous post).

Re Le Mans vs Rally car suspension: Just as with the progressive coupling of air cited above, believe it or not, maximum suspension compliance is very much a goal of Le Mans and even F1 chassis designers, as it allows the wheels to maintain contact with the road in bumpy corners (running out of suspension travel over a bump in a turn is precisely what killed Senna). This compliance however is sacrificed in the name of resisting the great aerodynamic down force these cars generate (literally thousands of pounds). Maintaining the vacuum under the car also requires holding fairly consistent ground clearance, which in turn requires firm suspension. Active suspension could solve all these issues, but it is not permitted.

jd*

Rerurn to Romy the Cat's Site